Control rod drop during hot zero power : RIA in BWR

Sammanfattning: During operation of nuclear power reactors reactivity initiated accidents (RIA) can occur, such as a control rod drop. If this occurs, the reactivity increase dramatically and leads to an increase in power, fuel enthalpy and fuel temperature. The fuel and reactor can be damaged. A methodology to simulate these accidents has been developed for Forsmark Nuclear Power Plant in cooperation with Westinghouse, referred to as the POLCA7 methodology. The POLCA7 methodology results in a limit for fuel failure regarding reactivity of the control rod that dropped in pcm/control rod percent. The limit is estimated from simulations in POLCA7, a static and deterministic code and POLCA-T, a dynamic code. The aim of this thesis is to evaluate the methodology and investigate what happens in a reactor if a control rod drops during hot zero power. Hot zero power is a phase during start-up, where the power is low (~2% of installed power) and the reactor have operation pressure and temperature. The POLCA7 methodology was applied on historic cycles in Forsmark. To evaluate the POLCA7 methodology the control rod drop was simulated in S3K, a dynamic software. The results from these cycles indicate that the limit for fuel failure set in the POLCA7 methodology in pcm/control rod percent is very conservative for fuel with low and medium burnup. Even though the limit is exceeded, the dynamic simulation in S3K shows that the fuel is far from failure regarding SSM limits in fuel enthalpy and cladding temperature. In this thesis new limits in POLCA7 has been generated, which is remarkably higher than the original limit from the POLCA7 methodology. To challenge the methodology, an unrealistic fuel design was simulated with fuel with high burnup surrounded by high reactive fuel. With this fuel design, the enthalpy limit from SSM was exceeded for the fuel with high burnup. The limit from the POLCA7 methodology was exceeded which indicate that the POLCA7 methodology meets the goal of detecting severe RIAs. Fuel with high burnup seems to be the most important fuel to investigate at a RIA simulation. Another discovery is that POLCA7 gives the most severe accident at 2% power, but in S3K it is given by 3-4% power. This is a problem with the POLCA7 methodology. Suggestions are made on how to lower the calculation time and improve the methodology. A control rod sequence that gives an even power distribution and a core with the fuel with high burnup in the periphery and only a few fresh fuels is preferred to avoid damage at a RIA. A control rod sequence was designed for the new cycle in Forsmark 1, in order to try to create a cycle without problems due to RIA. The new sequence was a success with no control rods exceeding the limit of 82 pcm/control rod percent, and it shows that conclusions about the impact of the sequence are correct. Conclusion is made that the methodology should be further investigated and there are good chances to develop a good and time efficient analysis in the future. One presented suggestion is to have a dynamic simulation of the incident instead of the axial simulation. The evaluation with SSM’s limits would then be direct.